-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathmethods.html
More file actions
424 lines (387 loc) · 20.9 KB
/
Copy pathmethods.html
File metadata and controls
424 lines (387 loc) · 20.9 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
<!DOCTYPE html>
<html lang="en">
<head>
<title>openMetaAnalysis</title>
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
<!-- jquery -->
<script src="//ajax.googleapis.com/ajax/libs/jquery/2.1.1/jquery.js"></script>
<script src="/openmetaanalysis.js"></script>
<script>
$(document).ready(function(){
//button handlers
$(".type2").on("click", function(){
if ($("#iqr").is(':checked')){
$("#subjects").css("background-color",lightgray);
}
else{
$("#subjects").css("background-color",white);
}
});
$("#bttncalculate").on("click", function(e){
e.preventDefault();
var subjects = parseFloat($("#denom1").val()) + parseFloat($("#denom2").val()) ;
if ($("#counts").is(':checked')){
weightedvalue = 100*((parseFloat($("#numer1").val()) + parseFloat($("#numer2").val()))/subjects);
//weightedvalue = (($("#numer1").val()*$("#denom1").val() + $("#numer2").val()*$("#denom2").val())/subjects);
weightedvalue = weightedvalue.toFixed(2).toString() + "%";
}
if ($("#percents").is(':checked')){
weightedvalue = (($("#numer1").val()*$("#denom1").val() + $("#numer2").val()*$("#denom2").val())/subjects);
weightedvalue = weightedvalue.toFixed(2).toString() + "%";
}
if ($("#means").is(':checked')){
weightedvalue = (($("#numer1").val()*$("#denom1").val() + $("#numer2").val()*$("#denom2").val())/subjects);
weightedvalue = weightedvalue.toFixed(2).toString();
}
$("#answer").html(weightedvalue);
});
$("#bttncalculate2").on("click", function(e){
e.preventDefault();
var width = Math.abs(parseFloat($("#upper").val()) - parseFloat($("#lower").val()) );
var se = 0;
if ($("#ci").is(':checked')){
var subjects = parseFloat($("#subjects").val());
se = width/2/1.96 //Chinn, 2000 PMID 11113947
sd = se * Math.sqrt(subjects) //Varies with sample size
}
if ($("#iqr").is(':checked')){
sd = width/1.35;
}
sd = sd.toFixed(2).toString();
$("#answer2").html(sd);
});
});
</script>
<style>
.wrapper
{
width: 860px;
margin-top: 0px;
margin-right: auto;
margin-bottom: 0px;
margin-left: auto;
}
table, td, th{
padding:10px;
border: 1px solid black;
border-collapse: collapse;
}
/* Menu from http://codepen.io/philhoyt/pen/ujHzd */
#primary_nav_wrap
{
margin-top:15px;
}
#primary_nav_wrap ul
{
list-style:none;
position:relative;
float:left;
margin:0;
padding:0
}
#primary_nav_wrap ul a
{
display:block;
background-color:#6DC6E7;
color:#FFFFFF;
text-decoration:none;
font-weight:bold;
//font-weight:700;
//font-size:12px;
line-height:32px;
padding:0 15px;
//font-family:"HelveticaNeue","Helvetica Neue",Helvetica,Arial,sans-serif
}
#primary_nav_wrap ul li
{
position:relative;
float:left;
margin:0;
padding:0
}
#primary_nav_wrap ul li.current-menu-item
{
background:#0022B4;
color:#FFFFFF;
}
#primary_nav_wrap ul a:hover
{
background:#0022B4;
color:#FFFFFF;
}
#primary_nav_wrap ul ul
{
display:none;
position:absolute;
top:100%;
left:0;
background-color:#6DC6E7;
color:#FFFFFF;
padding:0
}
#primary_nav_wrap ul ul li
{
float:none;
width:200px
}
#primary_nav_wrap ul ul a
{
line-height:120%;
padding:10px 15px;
}
#primary_nav_wrap ul ul ul
{
top:0;
left:100%
}
#primary_nav_wrap ul li:hover > ul
{
display:block
}
</style>
</head>
<body>
<!--
http://jquery-csv.googlecode.com/git/examples/basic-usage.html
http://jquery-csv.googlecode.com/git/examples/flot.html -> seems best and most stable without reticulating spines error
http://code.google.com/p/jquery-csv/wiki/API -> instructions
-->
<div class="wrapper">
<div class="page-header">
<div style="color:#0022B4">
<h1>openMetaAnalysis: <span style="font-size:50%">collaborative and continuous</span></h1>
</div>
<nav id="primary_nav_wrap">
<ul id="menu">
<li class="listitem"><a href="http://openmetaanalysis.github.io/" class="menuitem">Home</a></li>
<li class="listitem"><a href="#" class="menuitem">Editors</a>
<ul>
<li><a href="https://openmetaanalysis.ocpu.io/home/www/editor.html" class="menuitem">Intervention</a></li>
<li><a href="https://openmetaanalysis.ocpu.io/home/www/editor.html#instruct" class="menuitem">Intervention (instructional)</a></li>
<li><a href="https://openmetaanalysis.ocpu.io/home/www/editor-dx.html" class="menuitem">Diagnosis</a></li>
<li class="listitem"><a href="https://openmetaanalysis.ocpu.io/home/www/arrows.html" class="menuitem">Arrows Plot</a></li>
</ul>
</li>
<li class="listitem"><a href="http://openmetaanalysis.github.io/methods.html" class="menuitem">Methods</a></li>
<li class="listitem"><a href="http://openmetaanalysis.github.io/using_github.html" class="menuitem">Using GitHub</a></li>
</ul>
</nav>
</div>
<div style="clear:both"> </div>
<h1 style="color:#0022B4">Methods</h1>
<div>Methods are adapted from:</div>
<ul>
<li><a href="http://handbook.cochrane.org/">Cochrane Handbook</a></li>
<li><a href="http://www.gradeworkinggroup.org/">GRADE Working Group</a>. openMetaAnalysis, similar to meta-analysis in general, focuses on the intial 4 - 6 steps of the GRADE process (see Box 1 at <a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC428525/">BMJ, 2006</a>).</li>
<li><a href="http://www.bris.ac.uk/quadas/quadas-2/">QUADAS-2</a> (for meta-analyses of diagnostic tests)</li>
<li><a href="http://www.prisma-statement.org/">PRISMA</a> Statement (PMID: <a href="http://pubmed.gov/19622512">19622512</a>) and Checklist. PRISMA items 7 - 11 are substantially affected by using open and continuous updating.</li>
</ul>
<div>
<span style="font-weight:bold;font-size:130%">Selecton of studies</span>
<span style="font-weight:bold"> (PRISMA Item 6)</span>
</div>
<p>Meta-analysis of randomized controlled trials:</p>
<ul>
<li>Random allocation</li>
</ul>
<p>Meta-analysis of diagnostic test accuracy:</p>
<ul>
<li>Cross sectional or cohort design (not case controls)</li>
<li>Random or consecutive selection of patients</li>
<li>All patients received a reference standard</li>
</ul>
<div>More strict criteria for inclusion may be based on the <a href="https://acpjc.acponline.org/shared/purpose_and_procedure.htm">criteria by American College of Physicians Journal Club (ACPJC)</a>.</div>
<div> </div>
<div>
<span style="font-weight:bold;font-size:130%">Searching for studies</span>
<span style="font-weight:bold"> (PRISMA Items 7,8)</span>
</div>
<p>Searching starts with identifying a recent meta-analysis. Additional trials may be identified by using:</p>
<ul>
<li>Cited reference searches for studies that cite the recent systematic review or the highest impact study within the most recent meta-analysis.
<ul>
<li><a href="http://thomsonreuters.com/thomson-reuters-web-of-science/">Web of Science</a></li>
<li><a href="http://scholar.google.com">Google Scholar</a></li>
</ul>
</li>
<li>Manual searches of online textbooks</li>
<li><a href="http://www.ncbi.nlm.nih.gov/books/NBK3827/#pubmedhelp.Finding_articles_rel">PubMed's Find Related Data portlet</a></li>
<li>Manual searches of PubMed</li>
<li>For clinical trials
<ul>
<li>Search of the Cochrane Library</li>
<li>Clinical Trials registries with methods suggested by Glanville (<a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4076126/">PMC4076126</a>):
<ul>
<li><a href="https://clinicaltrials.gov/ct2/search/advanced">ClinicalTrials.gov</a> (for studies with completed results and the search term random*)</li>
<li><a href="http://apps.who.int/trialsearch/AdvSearch.aspx">International Clinical Trials Registry Platform (ICTRP)</a> (WHO)</li>
</ul>
</li>
</ul>
</ul>
<p>Please add studies to repositories if you find qualifying trials that have not been included.</p>
<div>
<span style="font-weight:bold;font-size:130%">Data abstraction</span>
<span style="font-weight:bold"> (PRISMA Items 9-11)</span>
</div>
<table style="border-style:none;border: none;">
<tr><td>
<div>Creating the files for PICO and Bias tables</div>
<ul>
<li><a href="http://openmetaanalysis.github.io/make-table-pico.html">http://openmetaanalysis.github.io/make-table-pico.html</a> will help make the content for the PICO tables.</li>
<li><a href="http://openmetaanalysis.github.io/make-table-bias.html">http://openmetaanalysis.github.io/make-table-bias.html</a> will help make the content for the Risk of bias table.</li>
</ul>
<div>Consider using an online collaborative text editor with a colleague to develop the xml files for the PICO and bias tables.</div>
<ul>
<li><a href="https://kobra.io">Kobra</a> is very easy to use. Kobra itself is collaboratively developed with <a href="http://www.firepad.io">Firepad</a>.</li>
</ul>
</td>
<td>
<form>
<table>
<caption>Calculation of population characteristics from 'Table 1' of studies for PICO Table in openMetaAnalysis</caption>
<tr><th>Group 1</th><th>Group 2</th></tr>
<tr><td><input type="text" id="numer1" placeholder="Proportion or mean"/></td><td><input type="text" id="numer2" placeholder="Proportion or mean"/></td></tr>
<tr><td><input type="text" id="denom1" placeholder="Number of subjects"/></td><td><input type="text" id="denom2" placeholder="Number of subjects"/></td></tr>
<tr><td colspan="2" align="center"><input id="counts" name="type" type="radio"><label for="counts">Counts</label> <input id="percents" name="type" type="radio"><label for="percents">Percentages</label> <input id="means" name="type" type="radio"><label for="means">Means</label> </td></tr>
<tr><td colspan="2"><button id="bttncalculate">Calculate weighted value</button> Weighted value: <span id="answer"></span></td></tr>
</table>
</form>
<form>
<table>
<caption>Calculation of standard deviation from confidence interval or interquartile range</caption>
<tr><td><input type="text" id="lower" placeholder="Lower limit"/></td><td><input type="text" id="upper" placeholder="Upper limit"/></td></tr>
<tr><td colspan="2" align="center"><input type="text" id="subjects" placeholder="Number subjects in group" style="width: 200px;" /></td></tr>
<tr><td colspan="2" align="center"><input id="ci" name="type2" class="type2" type="radio"><label for="ci">Confidence interval</label> <input id="iqr" name="type2" class="type2" type="radio"><label for="iqr">Interquartile range</label></td></tr>
<tr><td colspan="2"><button id="bttncalculate2">Calculate standard deviation</button> SD: <span id="answer2"></span></td></tr>
</table>
</form>
<div>Details about calculating standard deviation from: <a href="http://pubmed.gov/11113947">confidence interval</a> and <a href="http://handbook.cochrane.org/chapter_7/7_7_3_5_mediansand_interquartile_ranges.htm">interquartile range</a>.</div>
</td></tr>
</table>
<div><span style="font-weight:bold;font-size:130%">Assessing quality of individual studies</span>
<span style="font-weight:bold"> (PRISMA Items 12)</span>
</div>
<ul>
<li>Studies of intervention
<ul>
<li><a href="http://handbook.cochrane.org/chapter_8/table_8_5_d_criteria_for_judging_risk_of_bias_in_the_risk_of.htm">Individual criteria for judging risk of bias</a> in a trial using the '<a href="http://handbook.cochrane.org/chapter_8/8_5_the_cochrane_collaborations_tool_for_assessing_risk_of_bias.htm">Cochrane's Risk of bias tool</a>'.(1)
<ul>
<li>Attrition: consider judgment of 'Unclear risk" it no Consort flow diagram</li>
<li>Selective reporting: consider judgment of 'Unclear risk" it no access to original protocol at a online trial registry</li>
</ul>
</li>
<li><a href="http://handbook.cochrane.org/chapter_8/table_8_7_a_possible_approach_for_summary_assessments_of_the.htm">Aggregating individual criteria into a summary judgment about a trial</a> (see column "Within a study")</li>
</ul>
</li>
<li>Studies of diagnostic accuracy
<ul>
<li><a href="http://www.bris.ac.uk/quadas/quadas-2/">Four criteria for judging risk of bias from QUADAS-2</a> (see Risk of bias row in the Table guided by the detailed criteria in the rows above)</li>
</ul>
</li>
</ul>
<div><span style="font-weight:bold;font-size:130%">Statistical analysis</span>
<span style="font-weight:bold"> (PRISMA Items 13 - 16)</span>
</div>
<p>Analyses are done with online at <a href="https://www.opencpu.org/">OpenCPU</a> using <a href="http://cran.r-project.org/">R</a>. Editors are available for <a href="https://openmetaanalysis.ocpu.io/home/www/editor.html">randomized controlled trials</a> and <a href="https://openmetaanalysis.ocpu.io/home/www/editor-dx.html">diagnostic tests accuracy studies</a>.</p>
<ul>
<li>Studies of intervention
<ul>
<li>We use the random effects model with the inverse variance as implemented in the R package <a href="http://cran.r-project.org/web/packages/meta/">meta</a>.</li>
<li>The Knapp-Hartung method adjusts test statistics and confidence intervals to gives wider (more conservative) confidence intervals and is one of the methods suggested by Cornell et al.(2)</li>
<li>The continuity correction of Diamond is used.(3)</li>
<li>Summary measures for binary outcomes include odds ratio and relative risk. Measures for continuous outcomes are either mean differences or <a href="http://handbook.cochrane.org/chapter_12/12_6_interpreting_results_from_continuous_outcomes_including.htm" class="hastip" id="smd_definition">standardized mean differences</a>.</li>
<li>Subgroup analyses and meta-regressions are done as needed with the R package <a href="http://cran.r-project.org/web/packages/meta/">meta</a>. This includes investigating correlation of the control rate with the outcome
<ul><li><a href="http://pubmed.gov/24339017,21908417,15358396,9777687?dopt=abstract">PubMed citations</li></ul>
</li>
<li>Heterogeneity is assessed with I<sup>2</sup></li>
</ul>
</li>
<li>Studies of diagnosis
<ul>
<li>Hierarchial bivariate model (Reitsma, 2005. PMID <a href="http://pubmed.gov/16168343">16168343</a>) as implemented in the R package <a href="http://cran.r-project.org/web/packages/Metatron/">metatron</a> with AUC from the R package <a href="http://cran.r-project.org/web/packages/mada/">mada</a></li>
</ul>
</li>
</ul>
<div>
<a id="user-content-assessing-quality-of-a-group-of-studies" class="anchor" href="#assessing-quality-of-a-group-of-studies" aria-hidden="true"><span class="octicon octicon-link"></span></a>
<span style="font-weight:bold;font-size:130%">Assessing quality across a group of studies</span>
<span style="font-weight:bold"> (PRISMA Items 15)</span>
</div>
<p><a href="http://handbook.cochrane.org/chapter_12/table_12_2_b_factors_that_may_decrease_the_quality_level_of_a.htm">Factors</a> developed by the <a href="http://www.gradeworkinggroup.org/">GRADE Working Group</a> are below for assessing a group of studies in a meta-analysis. (4) Specific criteria for each factor are are based on those used by the <a href="http://back.cochrane.org/">Cochrane Back Group</a> with modifications noted below.(PMID: <a href="http://pubmed.gov/23362516">23362516</a>)</p>
<table>
<tbody><tr>
<th>Factors</th>
<th>Criteria</th>
</tr>
<tr>
<td>Limitations in the design and implementation of available studies</td>
<td>• Serious risk of bias: More than 25% of participants from studies with low methodological quality as measured by the Cochrane's (interventions) or QUADAS-2 (diagnostic tests) Risk of bias tool (see above)<br>• Very serious risk of bias: More than 50% of participants from studies with low methodological quality as measured by the Cochrane's (interventions) or QUADAS-2 (diagnostic tests) Risk of bias tool</td>
</tr>
<tr>
<td>Indirectness</td>
<td> </td>
</tr>
<tr>
<td>Heterogeneity or inconsistency of results<br>(modified from the Cochrane Back Group)</td>
<td>• Serious heterogeneity: >=75% of participants from studies with findings in the same direction or I<sup>2</sup> >= 30%.(30% to 60% may represent moderate heterogeneity per <a href="http://handbook.cochrane.org/chapter_9/9_5_2_identifying_and_measuring_heterogeneity.htm">Identifying and measuring heterogeneity</a>)<br>• Very serious heterogeneity: I<sup>2</sup> >= 60% as 50% to 90%: may represent substantial heterogeneity.(5)</td>
</tr>
<tr>
<td>Imprecision of results<br>(modified from the Cochrane Back Group)</td>
<td>• Serious imprecision: Fewer than 2000 participants for each outcome (PMID: <a href="http://pubmed.gov/11158556">11158556</a>) or confidence intervals that include clinically unimportant outcomes<br>• Very serious imprecision: Fewer than 300 participants for each outcome.(PMID: <a href="http://pubmed.gov/23362516">23362516</a>)
<div>An alternative approach to determining imprecision is to use the "optimal information size" (PMID: <a href="http://pubmed.gov/21839614">21839614</a>) with an online calculator (<a href="http://www.stat.ubc.ca/~rollin/stats/ssize/b2.html">http://www.stat.ubc.ca/~rollin/stats/ssize/b2.html</a>).</div>
</td>
</tr>
<tr>
<td>Probability of publication bias</td>
<td>This area of meta-analytic practice is evolving and presently only addresses studies of interventions. See discussion at <a href="http://handbook.cochrane.org/chapter_10/10_4_5_summary.htm">http://handbook.cochrane.org/chapter_10/10_4_5_summary.htm</a>.
<ul>
<li>If more than 10 studies are present, test for the small study effect with the Egger test for continuous outcomes or the Rucker test for binary outcomes (<a href="http://cran.r-project.org/web/packages/metasens/">CRAN</a> and PMIDs: <a href="http://pubmed.gov/17592831,19836925">17592831,19836925</a>).</li>
<li>When less than 10 studies are present, study size of less than 50 or 1000 patients total (PMID: <a href="http://pubmed.gov/23616031">23616031</a>) or 100 per arm (PMID: <a href="http://pubmed.gov/20639294">20639294</a>) in most of all available studies may suggest small study effect.</li>
</ul>
</td>
</tr>
</tbody></table>
<div> </div>
<div><span style="font-weight:bold;font-size:130%">Summary of Findings Table</span>
<span style="font-weight:bold"> (Not part of PRISMA checklist)</span>
</div>
Summary of Finding Tables (SoF) are detailed in the Cochrane Handbook (<a href="http://handbook.cochrane.org/chapter_11/11_5_summary_of_findings_tables.htm">Chapter 11</a>) and by Guyatt et al (PMID: <a href="http://pubmed.gov/21195583">21195583</a>). SoF are currently produced with <a href="http://tech.cochrane.org/revman/gradepro">GRADEpro</a>.
<div> </div>
<h3>References</h3>
<div>
<ol class="task-list">
<li><a href="http://handbook.cochrane.org/chapter_8/8_5_the_cochrane_collaborations_tool_for_assessing_risk_of_bias.htm">Risk of bias tool</a>. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from <a href="http://www.cochrane-handbook.org">http://www.cochrane-handbook.org</a>
</li>
<li>Cornell JE, Mulrow CD, Localio R, Stack CB, Meibohm AR, Guallar E, Goodman SN. Random-effects meta-analysis of inconsistent effects: a time for change. Ann Intern Med. 2014 Feb 18;160(4):267-70. PMID: <a href="http://pubmed.gov/24727843">24727843</a></li>
<li>Diamond GA, Bax L, Kaul S. Uncertain effects of rosiglitazone on the risk for myocardial infarction and cardiovascular death. Ann Intern Med. 2007 Oct 16;147(8):578-81. PMID: <a href="http://pubmed.gov/17679700">17679700</a></li>
<li><a href="http://handbook.cochrane.org/chapter_12/12_2_1_the_grade_approach.htm">The GRADE approach</a>. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from <a href="http://www.cochrane-handbook.org">http://www.cochrane-handbook.org</a>
</li>
<li><a href="http://handbook.cochrane.org/chapter_9/9_5_2_identifying_and_measuring_heterogeneity.htm">Identifying and measuring heterogeneity</a>. In: Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from <a href="http://www.cochrane-handbook.org">http://www.cochrane-handbook.org</a>
</li>
</ol>
</div>
<div>Click here to see the <a href="https://github.com/openMetaAnalysis/home/pulse">revision history</a> of the source code.</div>
<div> </div>
<script>
//For gh-pages
//Page history and edit
var pagename = location.pathname.split('/').slice(-1);
if (pagename.toString().length < 1){pagename = "index.html"}
document.write("<div style='text-align:center'><a href='https://github.com/openMetaAnalysis/openMetaAnalysis.github.io/blob/master/" + pagename + "'>Edit this page</a> - <a href='https://github.com/openMetaAnalysis/openMetaAnalysis.github.io/commits/master/" + pagename + "'>Page history</a></div>")
</script>
</div>
<script src="javascripts/scale.fix.js"></script>
<script type="text/javascript">
var gaJsHost = (("https:" == document.location.protocol) ? "https://ssl." : "http://www.");
document.write(unescape("%3Cscript src='" + gaJsHost + "google-analytics.com/ga.js' type='text/javascript'%3E%3C/script%3E"));
</script>
<script type="text/javascript">
try {
var pageTracker = _gat._getTracker("UA-56740469-1");
pageTracker._trackPageview();
} catch(err) {}
</script>
<div id="tip" style="display:none"></div>
</body>
</html>